[12:22] <mdz> elmo: permanently fixed meaning we need to find a new mirror?
[12:23] <elmo> no
[12:23] <elmo> meaning that it's setup in such a way that we can fix most problems ourselves and be more reactive when it breaks
[12:24] <elmo> we have plenty of other us.a.u.c candidates, so even if we did have to find another, it wouldn't be disastrouis
[12:29] <sivang> cool, Deer Park download upgrades and installs them automaticall, although noting all my previous ones are not compatible
[12:32] <sivang> mdz: "Under Consideration" in https://wiki.ubuntu.com/DapperGoals means is still not certain if ubuntu drivers want this specs to be targetted for dapper?
[12:38] <mdz> sivang: yes
[12:40] <crimsun> elmo: Thanks for the tapiir sync earlier. Would you please sync tagtool, synaesthesia, and sympa from Sid (ok to override Ubuntu changes)?
[12:45] <sivang> mdz: is there anything I can sort out for your decision re home-user-backup?
[12:45] <sivang> mdz: (is the spec incomplete, looks impossible, whatever?)
[12:46] <mdz> sivang: I haven't had a chance to read it yet
[12:47] <sivang> mdz: k, thanks and sorry again. <mental note>please be more patient</mental note> 
[12:50] <MarioMeyer> infinity, ping
[12:56] <dholbach> good night everybody
[12:56] <ogra> night dholbach 
[12:56] <dholbach> night ogra
[12:57] <mvo> good night
[12:57] <ogra> night mvo 
[12:58] <mvo> mdz: a last question (before I go to bed), do you need a debdiff or more information on the langauge-selector upload for breezy-updates?
[01:00] <mdz> mvo: a debdiff would be welcome
[01:00] <ogra> yay
[01:02] <wasabi_> Hmm. At some point weren't devices supposed to be accessable by a uuid in /dev?
[01:02] <mdke> i noticed the password dialogue for network-admin spells it "network-admins". Is that worthy of a bug?
[01:03] <mvo> mdz: http://people.ubuntu.com/~mvo/language-selector_0.1.1.debdiff (important bits are the first two)
[01:03] <crimsun> elmo: thanks!
[01:04] <mdz> mvo: the changelog looks incomplete
[01:04] <mdz> mdke: it's a bug that it uses the name of the executable at all, rather than a meaningful name
[01:05] <mdke> mdz, very true
[01:06] <mvo> mdz: ups, wrong file (sorry, late here). I updated it, please reload
[01:10] <mdke> mdz, is that to file under gksudo or gst?
[01:10] <mdz> mdke: gst
[01:11] <mdz> mvo: ok, thanks
[01:11] <mdke> mdz, ok thanks will do that
[01:11] <mdz> mdke: severity: minor please
[01:16] <mdke> mdz, indeed
[01:16] <mdke> mdz, it is dapper only anyhow, so i'm sure someone can fix it in the next 4 months or so
[01:16] <mdz> really?  I thought I noticed it in breezy as well
[01:17] <mdz> I forgot to file it though
[01:17] <mdke> yes, the executable name is in breezy, but not the typo
[01:26] <MarioMeyer> infinity, ping
[01:28] <infinity> MarioMeyer : pong.
[01:28] <MarioMeyer> :P
[01:28] <infinity> (Please ping via /msg in the future, I suck at watching channel chatter when I'm busy)
[01:28] <MarioMeyer> someone said you're the php5 maintainer
[01:28] <MarioMeyer> can i pvt u then?
[01:29] <infinity> I've heard that same rumour.
[01:29] <infinity> Yeah, msg away.
[02:15] <daniels> lamont-away: could you please have a look at why all the chroots are broken? check the glibc build log for an example
[02:15] <lamont> daniels: that'd be because gcc-4.0 build-depends libgcj6, which build-depends the current version of gcc-4.0
[02:15] <lamont> and I've been slowly working through analyzing how screwed that is
[02:16] <daniels> lamont: ouch, christ
[02:16] <daniels> lamont: okay, thanks :)
[02:16] <tseng> cool, something more broken than mono 1.0
[02:17] <ajmitch> if that is possible
[02:17] <lamont> daniels: basically, gettext Depends: po-debconf Depends: libgcj6, which is unbuildable until gettext is installable
[02:17] <daniels> and not my fault this time.  amazing.
[02:45] <mojo> I find an issue with GtkFileChooser but I need some confirmation, the issue is:
[02:46] <mojo> for Save FileChooser dialog, for example I am saving a zip file then, if the name of the file in 'Name:' are the same as the name of the folder in current folder, FileChooser automatically jump into that folder instead.
[02:47] <ptlo> mojo, i believe that's a feature, not a bug
[02:47] <mojo> ptlo: sure?
[02:48] <mojo> mojo: but that GtkFileChoose is a Save Type
[02:48] <mojo> mojo: if users want to choose Folder to save in, they must do it via extension folder choosing option
[02:53] <spstarr_home> hmm anyone report debtags segfaulting in today's build?
[02:53] <daniels> check bugzilla
[02:53] <spstarr_home> none recorded
[02:53] <daniels> that's where those sorts of reports tend to go
[02:53] <daniels> there's your answer, then?
[02:53] <spstarr_home> 0x0810c34e in pkgAcquire::Item::IsTrusted ()
[02:54] <spstarr_home> im debugging it
[02:54] <daniels> enjoy
[03:01] <jdong> Failed to fetch http://archive.ubuntu.com/ubuntu/dists/breezy/main/binary-i386/Packages.gz  MD5Sum mismatch
[03:01] <jdong> Failed to fetch http://archive.ubuntu.com/ubuntu/dists/breezy/multiverse/binary-i386/Packages.gz  MD5Sum mismatch
[03:01] <jdong> archive.ubuntu.com problems.... :-/
[03:03] <elmo> I doubt it
[03:03] <elmo> breezy's Packages file hasn't changed since it released
[03:04] <elmo> it's far more likely there's a broken http proxy between you and archive
[03:07] <jdong> elmo: yep, my bad
[03:07] <jdong> damn squid server
[03:08] <tseng> exclude a.u.c
[03:08] <jdong> yep
[03:08] <tseng> caching packages.gz seems to always be a Bad Thing
[03:08] <jdong> or at least Release, Packages.gz files
[03:09] <tseng> even apt-proxy seems to have a penchant for spectacuarly breaking the cached files
[03:09] <jdong> yeah
[03:09] <ajmitch> even apt-proxy?
[03:09] <ajmitch> that's apt-proxy's main feature
[03:10] <tseng> i didnt mean to impune its tremendous usefulness
[03:10] <ajmitch> I have no problems doing so
[03:18] <spstarr_home> ca.archive.ubuntu.com is missing its gpg key
[03:19] <spstarr_home> b/n
[03:19] <spstarr_home> n/m
[03:24] <jdong> hey, how's the FF RC3 import
[03:24] <jdong> I'm gonna build it for fun...
[04:04] <wasabi> wasabi@kyoto:~$ ldd /usr/bin/epiphany  | grep gtkem
[04:04] <wasabi>         libgtkembedmoz.so => not found
[04:04] <wasabi> Somehow this situation has been allowed.
[04:22] <jdong> wasabi: yeah, sloppy merge job for FF rc2... being fixed
[04:23] <wasabi> k
[04:24] <jdong> wasabi: it seems like it's fixed in rc3's upload, though a recompile of epiphany is necessary
[04:25] <jdong> jdong@shuttle:~$ ldd /usr/bin/epiphany  | grep gtkem
[04:25] <jdong>         libgtkembedmoz.so => /usr/lib/mozilla-firefox/libgtkembedmoz.so (0xb7ef8000)
[04:25] <jdong> yep
[04:25] <jdong> recompile epiphany
[04:25] <wasabi> checking whether we have a gtk 2 gecko build... configure: error: This program needs a gtk 2 gecko build
[04:26] <wasabi> I don't have libgtkmozembed there. ;)
[04:26] <jdong> version 1.4.99+1.5rc3.dfsg-1ubuntu2
[04:26] <jdong> not compiled yet
[04:26] <jdong> apt-source it
[04:26] <wasabi> Ahh
[04:27] <wasabi> k
[04:27] <wasabi> Didn't see it on changes.
[04:27] <wasabi> But I see it now
[04:27] <jdong> needs force-overwrite to install, just a heads-up
[04:27] <wasabi> k
[04:28] <wasabi> thanks. ;0
[04:28] <jdong> np
[04:28] <jdong> actually researching into this for backports... not a dapper user yet
[04:29] <wasabi> Guess nobody likes ~ in versions yet.
[04:29] <jdong> lol
[04:29] <jdong> IMO *-updates should use it
[04:29] <jdong> security could even benefit
[04:29] <jdong> oh well, not my call :)
[04:29] <daniels> er, how would security benefit?
[04:30] <daniels> given that the only thing that ever gets backported is firefox
[04:30] <jdong> I meant the -security repo...
[04:30] <jdong> daniels: the mozilla naming blunder?
[04:30] <jdong> late hoary
[04:30] <jdong> that was fixed in BREEZY-updates?
[04:30] <wasabi> I don't know about any of that.
[04:31] <wasabi> I just think it looks cleaner.
[05:24] <Amaranth> anyone looked into the supposed autoconfiguration of Xorg that nexenta does?
[05:28] <daniels> they just use Xorg -configure
[05:28] <daniels> which I think is a tremendously bad idea
[05:28] <daniels> but, y'know, more power to them
[06:00] <Amaranth> daniels: Wasn't autoconfiguration supposed to be a feature of 7?
[06:01] <daniels> nope
[06:01] <daniels> a whole bunch of us have plans to make it workable later on
[06:03] <Amaranth> using hal and such?
[06:03] <daniels> yeah
[06:03] <daniels> with hotplug input devices
[06:03] <daniels> hotplug displays later on down the track
[06:03] <Amaranth> that would be _awesome_
[06:03] <daniels> we'll drag X into the 1990s if it kills us
[06:04] <dilinger> and by "us", you mean we users as well, right? :)
[06:05] <daniels> i can't imagine there being another breezy
[06:09] <wasabi> =( epiphany crashes with new ff. alas.
[06:09] <Burgundavia> daniels, not a fan of unmaintainable crap?
[06:10] <Burgundavia> desrt, have you raised our concerns about g-p-m upstream on desktop-devel?
[06:10] <desrt> negative.  martin matthew and myself came to an agreement
[06:10] <desrt> looks like g-p-m is in
[06:10] <desrt> although i DID get an interesting email today
[06:10] <desrt> not sure what to make of it
[06:11] <Burgundavia> from who?
[06:11] <desrt> there's this thing called kpowersave
[06:11] <desrt> [some guy] 
[06:11] <Burgundavia> desrt, in for dapper or in for .14?
[06:11] <desrt> and he kept writing itl ike (k)powersave
[06:11] <desrt> implying that the k is somehow optional
[06:11] <desrt> Burgundavia; dapper
[06:11] <desrt> Burgundavia; honestly i'm confused and possibly annoyed
[06:11] <Burgundavia> hm
[06:11] <Burgundavia> http://sourceforge.net/projects/powersave/
[06:11] <desrt> he emailed me at desrt@cas.mcmaster.ca which is quite odd
[06:12] <daniels> desrt: feel free to ...
[06:12] <desrt> From: Michael Biebl <biebl@teco.edu>
[06:13] <Burgundavia> well, that powersave I just mentioned does seem to fit the bill of what you describeed
[06:13] <Burgundavia> it has a WM applet
[06:13] <desrt> he's doing odd things like suggesting that we ask the user confusing questions about their laptop's power capabilities during install
[06:13] <Burgundavia> doesn't seem like it uses HAL yet
[06:13] <Burgundavia> http://powersave.sourceforge.net/Todo.html#Todo
[06:14] <daniels> Burgundavia: the unmaintainable crap wasn't my plan; it was more about the horrendous world-ending breakage that users felt
[06:14] <desrt> daniels; my damn video card still doesn't work </complaint>
[06:14] <desrt> *ahem*
[06:15] <desrt> i suspect, actually, it has nothing to do with drivers.  i think my MTRRs are wonked (however that works)
[06:15] <daniels> if you have mtrr issues, it's the kernel's fault
[06:15] <desrt> this is what i believe.
[06:16] <daniels> if you're waiting for r300, you'll have to wait for gcc to be fixed now that glibc's fixed to work with new binutils
[06:16] <desrt> i am waiting for 2615
[06:16] <daniels> and then the kernel also needs new drm for pcie r300
[06:16] <desrt> agp r350 here
[06:16] <Burgundavia> same
[06:16] <desrt> RV350 even
[06:17] <Burgundavia> desrt, I don't know enough about the internals to know if (k)powersave is sane
[06:17] <Burgundavia> but the UI is total crack, I can tell you taht
[06:18] <desrt> Burgundavia; it's a problem of "too big"
[06:18] <desrt> it's practically official.  i'm dead weight for dapper
[06:18] <desrt> i'm in the last 6 months of my undergraduate degree
[06:19] <desrt> so i'll see y'all for dapper+1
[06:19] <desrt> and i want to do it right then
[06:19] <Burgundavia> ouch
[06:19] <desrt> but now we don't have enough time.  mjg59 wants g-p-m and i'm cool with that
[06:19] <desrt> it's an acceptable stop-gap solution and he's pushing really hard for it
[06:20] <desrt> it also has the interesting feature of cutting me almost entirely out of the loop... which i would normalyl be unhappy about but due to time constraints elsewhere... eh
[06:21] <desrt> i'm going to keep this email around.  using/fixing/cleaning-up/whatever powersave might be an appropriate thing to do for dapper+1?
[06:21] <Burgundavia> sounds good
[06:22] <Burgundavia> I like the getting richard and this Kpowersave guy to get together and agree on a common backend
[06:23] <desrt> more like: get all the stakeholders together and get them to agree on what the problem is
[06:23] <Burgundavia> ya
[06:23] <Burgundavia> and some good common points
[06:24] <desrt> richard does -not- want to hear "system daemon"
[06:24] <daniels> argh, you said stakeholders
[06:24] <desrt> i emailed him about it and it's been on the g-p-m mailing lists a few times
[06:24] <Burgundavia> ah, elmo deleted RideTheWildElmo
[06:24] <daniels> maybe they could all leverage synergies going forward
[06:25] <wasabi> And vertical markets.
[06:25] <Burgundavia> in a way that provides value to all
[06:25] <desrt> daniels; only if we want to have a paradigm shift to a new phase of emerging technology
[06:25] <daniels> https://wiki.ubuntu.com/RideTheWildElmo?action=info
[06:26] <desrt> hahahah
[06:26] <Burgundavia> James is probably going to block me from editing the wiki now
[06:26] <desrt> jeff seems to be having a bit of a fight with him here
[06:26] <desrt> Burgundavia; i do believe you are obliged to add your name to the list
[06:27] <fabbione> morning
[06:27] <desrt> g'morn
[06:27] <desrt> anyway... seriously though... there are stakeholders and richard is probably one of them
[06:28] <Burgundavia> he has spent the most time on it so far
[06:28] <desrt> and he doesn't want to hear "system daemon" because it pretty much means "toss your code in the bin -- time for a rewrite"
[06:28] <Burgundavia> ya
[06:28] <desrt> which is fair
[06:28] <desrt> but not objective
[06:29] <Burgundavia> any idea how gnome is going to move on this?
[06:29] <desrt> no.
[06:30] <desrt> maybe i should make my opinion known to gnome.... but honestly, i really don't want to shut richard down
[06:30] <desrt> he has done a lot of work
[06:30] <ajmitch> Burgundavia: you revived that page?
[06:30] <desrt> writing a "why gnome should not use g-p-m" email to d-d-l for example would make me feel like a colossal asshat
[06:31] <Burgundavia> ajmitch, Good Urls never die
[06:31] <ajmitch> but naughty wiki gnomes do
[06:32] <desrt> a threat? :)
[06:32] <ajmitch> not from me :)
[06:32] <Burgundavia> I can handle being locked out of the wiki
[06:32] <desrt> but can the wiki handle it? <tear>
[06:32] <ajmitch> elmo is probably already on his way to your house to kill you
[06:33] <Burgundavia> good thing I have a rabid (4" tall) dog here to guard me
[06:33] <ajmitch> those ones are the worst
[06:37] <Burgundavia> hmm, d-d-l discussion of g-p-m sort of ran out of steam discussing battstat and interacting with gpm
[06:38] <desrt> oh.  that's the other thing about g-p-m that bugs me :p
[06:38] <desrt> it's reinventing the battstat wheel
[06:39] <Burgundavia> so would have battstat for the actual display, and g-p-m for the configuration dialog and the actual work?
[06:40] <desrt> i guess, but i don't know
[07:55] <fabbione> hmmm
[07:55] <fabbione> daniels: did you upload the last xterm?
[07:56] <daniels> \sh maintains it
[08:11] <linnuxxy> how to get the ifconfig information in C?
[08:11] <Chipzz> wrong channel
[08:46] <siretart> morning ubuntu devs!
[08:46] <siretart> did the c2a transition already start?
[08:49] <siretart> hey Scott!
[08:49] <Keybuk> morning
[08:50] <fabbione> siretart: nope
[08:50] <fabbione> siretart: the archive is unbuildable at the moment
[08:50] <Amaranth> siretart: i guess, latest dapper dist-upgrade uninstalls a bunch of things thanks to libstdc++6
[08:50] <fabbione> gcj-4.0 transition need to be done first
[08:50] <pitti> Hi siretart 
[08:51] <fabbione> hey Keybuk 
[08:53] <siretart> fabbione: morning. and thanks for explanation
[08:53] <Mithrandir> elmo: pong
[08:53] <siretart> huhu pitti 
[09:35] <sivang> Morning all
[09:50] <siretart> fabbione: any reason why ubuntu's iptables does not ship /lib/iptables/libipt_recent.so?
[09:51] <fabbione> siretart: dunno
[09:51] <fabbione> pitti: thanks for fixing my bugs :)
[09:51] <Keybuk> siretart: it's built-in to libc isn't it?
[09:51] <fabbione> pitti: krb4 ;)
[09:51] <Keybuk> siretart: what symbols are you looking for?
[09:52] <siretart> Keybuk: I'm trying to use the 'recent' match from iptables
[09:52] <pitti> fabbione: it also bothered me a lot
[09:52] <Keybuk> that's not a symbol :)
[09:52] <siretart>  iptables v1.3.1: Couldn't load match `recent':/lib/iptables/libipt_recent.so: cannot open shared object file: No such file or directory
[09:52] <hunger> Anyone having problems with debtags? postinst reports a segfault for me.
[09:52] <fabbione> pitti: ok.. 
[09:52] <fabbione> hunger: known problem
[09:52] <hunger> fabbione: Good, I can't bugreport from here;-)
[09:52] <pitti> fabbione: and it's part of the ReducingDuplicates killing rave :)
[09:52] <siretart> Keybuk: I didn't talk about symbols, but this libipt_recent.so file
[09:53] <fabbione> pitti: even better ;)
[09:53] <Keybuk> ah, sorry, I'm confusing those things with a kernel syscall interface
[09:53] <siretart> fabbione: okay, I'll file a bug in bugzilla then, okay?
[09:54] <fabbione> siretart: iptables in dapper is FTBFS.. so let's get that one fixed first and see
[09:54] <fabbione> siretart: but if you want to open a bug, just go ahead
[09:54] <fabbione> i don't maintain iptables anyway ;)
[09:54] <hunger> Is there a "ReducingTransitionalPackages" killing rave as well?
[09:54] <siretart> fabbione: you was the last uploader. that the reason I bug you ;)
[09:55] <siretart> well, the one mentioned in /u/s/d/iptables/changelog.Debian.gz, even
[09:55] <fabbione> am I?
[09:56] <siretart> you uploaded 1.3.1-2ubuntu1
[09:56] <siretart> thats the version in breezy
[09:56] <fabbione> hmm crap
[09:56] <siretart> ;)
[09:56] <fabbione> sucks to be me
[09:56] <fabbione> ok file a bug, but let bugzilla assign it to who is responsible for it
[09:56] <fabbione> there might be a default assignee ;)
[09:57] <Keybuk> siretart: there's no documented reason for it, I suspect it's just because Debian didn't
[09:57] <siretart> Keybuk: I'll investigate it
[09:59] <siretart> merge seems borked. I'm on it
[10:00] <fabbione> siretart: yes it is
[10:01] <Keybuk> ya know, this responsibility-by-first-letter thing is getting scary
[10:01] <Keybuk> seb just got every merge beginning 'g' assigned to him
[10:01] <Keybuk> jr everything beginning 'k'
[10:01] <Keybuk> and chmj everything beginning 'j'
[10:02] <seb128> pitti: hum
[10:02] <seb128> $ grep DB_INIT_TXN * -r
[10:02] <Treenaks> Keybuk: that's all ubuntu-specific, no merge required
[10:02] <seb128> calendar/libical/src/libicalss/icalbdbset.c:  flags = DB_INIT_LOCK | DB_INIT_TXN | DB_CREATE | DB_THREAD | 
[10:02] <pitti> seb128: shit
[10:02] <Keybuk> oh, and doko has everything beginning 'p' :)
[10:03] <seb128> pitti: 
[10:03] <seb128> $ grep DB_TXN * -r | grep -v libdb
[10:03] <seb128> calendar/libical/src/libicalss/icalbdbset.c:  DB_TXN *tid;
[10:03] <seb128> calendar/libical/src/libicalss/icalbdbset.c:int icalbdbset_acquire_cursor(DB *dbp, DB_TXN *tid, DBC **rdbcp) {
[10:03] <seb128> calendar/libical/src/libicalss/icalbdbset.c:  DB_TXN *tid;
[10:03] <seb128> ...
[10:03] <seb128> etc
[10:04] <siretart> responsibility by first letter? is this a canonical internal assignment or did I miss something?
[10:04] <pitti> seb128: that means that 4.1, 4.2 and 4.3 are incompatible then
[10:04] <seb128> :(
[10:05] <Keybuk> siretart: a long-running joke.  seb128 once disavowed responsibility for a bug because it was in a package which began 'x' and he only deals with 'g' :)
[10:05] <fabbione> ehhehe
[10:06] <siretart> lol
[10:06] <siretart> ;)
[10:06] <seb128> which turned to not be a good idea
[10:06] <seb128> doko tried to give me gcc :p
[10:06] <hunger> seb128: You should have picked a letter not kontaining most of gnome;-)
[10:06] <seb128> and jbailey glibc :p
[10:08] <Keybuk> did you take them?
[10:09] <seb128> you don't want me to take them :)
[10:09] <seb128> I've enough to break with GTK :p
[10:10] <Keybuk> I use Kubuntu now (*)
[10:10] <Keybuk> (* note: lie)
[10:10] <chmj> huh ? 
[10:13] <Amaranth> it's not fun being a gnome/gtk guy when people start showing you all the cool qt things
[10:13] <Amaranth> like being threadsafe
[10:13] <seb128> trooooolll
[10:13] <dholbach> hellas
[10:13] <seb128> hi dholbach
[10:14] <dholbach> hey seb :)
[10:14] <Amaranth> troll? i'm just pissed off at threads :P
[10:14] <Keybuk> don't use threads
[10:14] <Nafallo> morning
[10:14] <Amaranth> yeah...
[10:15] <Amaranth> what else am i supposed to use? people always tell me to use twisted
[10:17] <siretart> Keybuk: is it possible that MoM does not deal at all with changes in binary files?
[10:17] <Keybuk> it's not only possible, it's true
[10:18] <zyga> morning
[10:18] <Keybuk> fortunately dpkg-source doesn't deal with changes in binary files either
[10:18] <siretart> okay. then I found the reason why the iptables merge is borked
[10:18] <Keybuk> MoM does get very confused with a couple of strange source formats though, usually native .tar.gz which an .orig.tar.bz2 inside it and patches
[10:18] <siretart> I'm testing my merge right now..
[10:19] <siretart> iptables includes the 'real' upstream tarball along with several patch-o-matic sources
[10:19] <Keybuk> we've never merged iptables?
[10:19] <siretart> there was no reason for, we didn't diverge since short before breezy release
[10:19] <Keybuk> ah, d'oh, I looked at the breezy source package
[10:20] <Keybuk> yeah, looking at that package, it's the kind of package MoM sometimes screws up on
[10:20] <Keybuk> which is why we have humans who can review them
[10:20] <Keybuk> but the humans screw up on those too
[10:20] <siretart> of course. 
[10:20] <Keybuk> which, imo, is pretty much proof that people shouldn't package things like that, damnit! :p
[10:20] <siretart> but it is a great tool in assisting merges
[10:21] <siretart>   debhelper: Depends: po-debconf but it is not going to be installed
[10:21] <siretart> thats a nice one :/
[10:22] <\sh> yeah
[10:23] <\sh> siretart: but it's libgcj6 which has problems with gcc4 base or something like this
[10:23] <fabbione> siretart: that's still the same as i told you before
[10:24] <siretart> fabbione: now I see that in action :)
[10:24] <fabbione> basically ~85% of the archive cannot be built atm
[10:24] <fabbione> if not more
[10:24] <Nafallo> simpified, the dapper archive is more or less corrupt? :-P
[10:24] <Nafallo> s/p/pl/
[10:25] <siretart> not corrupt. simply broken
[10:25] <fabbione> the archive is good, just not buildable :)
[10:25] <Nafallo> so not broken, but corrupt ;-)
[10:26] <Nafallo> i.e. a little bit broken
[10:26] <Nafallo> :-)
[10:28] <Kamion> no, corrupt would indicate malice
[10:29] <Kamion> changes not corresponding to what's been uploaded
[10:29] <Nafallo> hm, oki.
[10:37] <pitti> seb128: which part of gnome starts the dbus session bus?
[10:37] <Treenaks> pitti: /etc/X11/Xsession.d/*
[10:37] <seb128> pitti: xorg
[10:38] <Treenaks> (I guess)
[10:38] <seb128> pitti: what Treenaks said
[10:38] <pitti> thanks
[10:38] <seb128> pitti: /etc/X11/Xsession.d/75dbus-1-utils_dbus-launch
[10:42] <pitti> seb128: a friend of mine just phoned me, he has no session dbus
[10:42] <pitti> seb128: he does not have dbus-1-utils installed; we should update the package description
[10:45] <pitti> Mithrandir: any idea why heimdal (which is supposedly krb5) depends on kerberos4kth-dev, which is krb4?
[10:45] <seb128> pitti: he doesn't use Ubuntu? Because ubuntu-desktop Depends on it ...
[10:45] <pitti> this kerberos mess drives me crazy...
[10:45] <fabbione> http://www.f-secure.com/weblog/archives/most_people_shirt.jpg
[10:45] <pitti> seb128: he does use Ubuntu, but he removed -desktop to get rid of all the stuff he does not need
[10:45] <fabbione> pitti: ^^
[10:46] <Nafallo> lol
[10:46] <pitti> fabbione: LOL, I read that quote a few days ago. Somebody should hit that guy very hard 
[10:46] <fabbione> ehehhe
[10:47] <\sh> I want to have this shirt please send :)
[10:47] <Treenaks> \sh: www.cafepress.com
[10:48] <\sh> hmmm....I'll ask amu...the quality is much better for the shirts ;)
[10:49] <pitti> Mithrandir: ah, nevermind, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=315059
[10:50] <pitti> torkel: ping
[10:50] <torkel> pitti: pong
[10:51] <pitti> torkel: just read your bug followup - you know a bit about krb4 vs. 5?
[10:51] <torkel> pitti: a bit
[10:52] <pitti> torkel: does heimdal 0.7.1 break any other apps?
[10:52] <pitti> torkel: i. e. would it be safe to get the experimental package into dapper?
[10:52] <\sh> "Most Companies don't know who their customers are, why should we care about them?"
[10:55] <Nafallo> why does my XChat default to show of URLs in w3m?
[10:55] <torkel> pitti: it shouldn't. You might have to rebuild som packages though (libsasl probably)
[10:55] <infinity> It might default to sensible-browser
[10:56] <pitti> torkel: right, I just updated cyrus-sasl2, and I need to rebuild it anyway to completely drop krb4 dependencies
[10:56] <pitti> torkel: but for that I need heimdal without krb4
[10:56] <pitti> Hi infinity 
[10:56] <torkel> pitti: I think it is safe to bring in the experimental package
[10:56] <Nafallo> oh!
[10:56] <Nafallo> infinity: x-www-browser is a broken symlink :-P
[10:57] <infinity> Nafallo : I assume firefox isn't properly updating the alternative.
[10:57] <torkel> pitti: and better to do it now, than just before :-)
[10:58] <pitti> right
[10:58] <Nafallo> Diziet: could firefox please update x-www-browser to /usr/bin/firefox instead of /usr/bin/mozilla-firefox? :-)
[10:58] <torkel> pitti: and most of the stuff in main (evolution, etc) are built against MIT anyway, iirc
[10:58] <infinity> Nafallo : Ouch, looks like breakage from the mozilla-firefox -> firefox transition.  Does that mean it's also broken in breezy, or did we have a compatibility link in breezy's firefox?
[10:59] <Nafallo> infinity: the latter I think. I noticed it with the 1.5-ish version :-)
[10:59] <infinity> Nafallo : Ahh, phew.  No problem then.  I'd recommend filing a bug if Diziet doesn't get back to you on IRC in the next 5 minutes.  IRC pings get lost.
[11:00] <Nafallo> infinity: yea, I will :-)
[11:00] <torkel> pitti: the only thing I can think of is how to handle those running a kthkrb4 KDC. If there are any...
[11:01] <pitti> torkel: well, we want to kill the ancient krb4 completely anyway
[11:03] <infinity> The odds that anyone would be running a krb4 KDC on Ubuntu are pretty freakin' slim, given how new Ubuntu is, and how old and useless krb4 is.
[11:04] <infinity> The only people who would be doing so would almost certainly be doing so with very good reason, know exactly what they're doing, and know how to fix their own stuff when we shoot their feet for them.
[11:04] <pitti> torkel: ok, I checked, the only package that needs a rebuild is cyrus-sasl2; that's easy
[11:04] <infinity> (IMO)
[11:05] <torkel> infinity: I couldn't agree more. I just wanted to hear it from someone else :-)
[11:05] <pitti> infinity: I'd test the new heimdal now and get rid of krb4. Unless you already planned to do the killrave yourself :)
[11:05] <infinity> I'll let you have the honour. :)
[11:05] <infinity> krb4 begone.
[11:05] <pitti> ok :)
[11:06] <torkel> yay :-)
[11:06] <infinity> I need to either stop working (it's 9pm) or sort out the libgcj6 mess.
[11:06] <infinity> I haven't decided yet if I value archive consistency or my own sanity more right now.
[11:06] <fabbione> infinity: the former > the latter
[11:06] <fabbione> doko can fix it when he wakes up
[11:06] <infinity> He can upload fixed packages, but they won't build.
[11:07] <infinity> So it'll stay broken until after I wake up again.
[11:07] <infinity> \o/
[11:07] <infinity> Hence my waffling and thinking I may just fix it later.
[11:09] <maswan> yay! death to krb4!
[11:09] <sivang> Nafallo: ask seb128 , he's already going to fix one gconf key for that for gnome-open :)
[11:10] <Nafallo> sivang: ?
[11:10] <pitti> I'm in a killing mood today and don't have any urgent security stuff, so why not do it today :)
[11:11] <Nafallo> pitti: you could always check MainInclusionReportLogcheck if you have nothing to do ;-)
[11:11] <infinity> pitti : If you want low-hanging fruit, apt-cache rdepends libssl0.9.7 and rebuild the world.
[11:11] <infinity> pitti : But none of it will BUILD until I fix libgcj6.
[11:11] <pitti> infinity: I though you wanted to use a clever script for that?
[11:12] <pitti> infinity: why does this java stuff break the whole archive?
[11:12] <infinity> gettext depends on it.
[11:12] <infinity> po-debconf depends on gettext.
[11:12] <pitti> on JAVA_
[11:12] <pitti> ???
[11:12] <sivang> Nafallo: s/mozilla-firefox/firefox/ issues :)
[11:13] <Nafallo> sivang: naah. that's firefox job to update :-)
[11:13] <seb128> this one is a firefox one
[11:13] <seb128> not for me
[11:13] <seb128> grumpf, just the right timing to screw debhelper
[11:14] <seb128> Ian broken all the GNOME stuff using firefox with his 1.5 upload
[11:15] <pitti> infinity: enjoy :) and thanks for fixing it soon
[11:15] <mahangu> anybody know why the dapper torrent for i386 is so slow?
[11:16] <siretart> mahangu: perhaps not enough seeds?
[11:16] <mahangu> siretart, yeah i figured, but how come?
[11:16] <siretart> no idea
[11:17] <mahangu> ok thanks
[11:18] <\sh> grmpf..why is tomboy not in the kde menus?
[11:20] <\sh> because it's not installed yay
[11:26] <mahangu> !javadebs
[11:27] <mahangu> !javadeb
[11:27] <mahangu> oh sorry
[11:27] <mahangu> wrong channel :S
[11:28] <siretart> there are javadebs?!
[11:28] <siretart> whats that?
[11:29] <mahangu> sorry
[11:30] <mahangu> that's a channel bot trigger in #ubuntu and #ubuntu-offtopic
[12:15] <hunger> Kamion: How are you progressing with the hotplug->udev trtansition?
[12:16] <ogra> hunger, thats Keybuk, not Kamion 
[12:16] <hunger> Keybuk: How are you progressing with the hotplug->udev trtansition?
[12:16] <hunger> ogra Thanks. I'm terrible at remembering names (esp. without faces to connect them to).
[12:17] <pitti> hunger: there is a slight chance that the next conf will be in .de :)
[12:17] <Keybuk> \o/  finally!  just got a call from alienware, my computer's gonna be shipped soon
[12:17] <Keybuk> yay
[12:18] <ogra> Keybuk, an alienware lappie ? 
[12:18] <Keybuk> hunger: well.  was there anything in particular you were interested in
[12:18] <Keybuk> ogra: no, new desktop
[12:18] <hunger> pitti: That would be neat.
[12:18] <ogra> ah, ok...
[12:18] <hunger> Keybuk: The deb so that I can depend on it;-)
[12:18] <ogra> i'd envy you if it was the new slimline lappie :)
[12:18] <Keybuk> hunger: why do you need to depend on it?
[12:18] <hunger> Keybuk: Xen has a couple of hotplug scripts I want to port.
[12:19] <Keybuk> we should probably look those over
[12:19] <Keybuk> they might not need to depend on the new udev, depending how complex they need to be
[12:19] <Keybuk> the current udev supports RUN rules
[12:19] <Keybuk> ie. /etc/udev/rules.d/85-hdparm.rules on dapper
[12:20] <hunger> Keybuk: Oh, then I can play with that already. Nice.
[12:21] <hunger> Keybuk: I have only glanced over the Xen scripts so far. When I was about to do that I heared you say that hotplug is going to go away.
[12:21] <hunger> Keybuk: So I decided to postpone that.
[12:24] <hunger> Keybuk: The script looks rather simple, shouldn't be a problem to port it to udev.
[12:28] <Keybuk> yeah, mostly you should probably move it to /lib/udev and run it with something like ACTION=="add", ..., RUN+="/lib/udev/SCRIPT"
[12:28] <Keybuk> where the ... qualifies when to run it
[12:28] <Keybuk> usually it's something like SUBSYSTEM=="blah" or whatever
[12:33] <ogra> Kamion, elmo ? who is processing the launchpad approvals from yesterdays CC meeting ? 
[12:33] <Kamion> ogra: I did them during the meeting ...
[12:34] <ogra> Kamion, MauricioHernandez isnt moved to the Active list yet
[12:34] <Kamion> ogra: he wasn't approved because mako didn't come back to give his ack
[12:34] <\sh> Kamion: mako came back...
[12:34] <Kamion> unless I missed that
[12:35] <\sh> he wanted to read the ircbacklog :=
[12:35] <ogra> oh, right... my fault, sorry for bothering
[12:35] <Kamion> ok then, he came back but didn't give his ack :)
[12:35] <\sh> btw..do I need to bother anyone to use my ubuntu.com address for uploads?
[12:35] <Nafallo> \sh: nope
[12:35] <tseng> \sh: it doesnt care about your address
[12:36] <tseng> \sh: just your key
[12:36] <\sh> or is it enough when it's in my gpg key as uid
[12:36] <tseng> the address is completely irrelevant
[12:36] <Kamion> it doesn't care whether the address is a uid on your key or not
[12:36] <tseng> i can upload slomo's changelog signed by me
[12:36] <\sh> ah ok :)
[12:39] <\sh> chmj: ping
[12:39] <\sh> chmj: are u working on http://bugzilla.ubuntu.com/show_bug.cgi?id=19248 (libquicktime merge) or can I reassign it to me...because I need it for another package
[12:50] <chmj> \sh: umh.. not currently 
[12:50] <\sh> chmj: so you don't mind if I take this over? :)
[12:54] <chmj> \sh: no, I don't 
[12:55] <\sh> chmj: k..thx
[12:56] <ogra> mjg59, did you see http://www.emperorlinux.com/images/contrib/acpi.png ?
[12:56] <ogra> it says in its README: ...eventually be the real application. This version has only been tested
[12:56] <ogra> 	on Fedora 4 and Ubuntu (Hoary & Breezy).
[12:56] <ogra> :)
[01:27] <\sh> grmpf
[01:28] <\sh> libquicktime transition
[01:29] <\sh> libquicktime1 is renamed to libquicktime0 
[01:30] <Keybuk> isn't that going the wrong way?
[01:30] <infinity> For libquicktime, it's proabbly the right way.
[01:30] <infinity> "Oh wait, this really IS an unstable API/ABI, fuck"
[01:30] <Keybuk> heh
[01:31] <infinity> Easier to give up and go to 0, than to increment the SOVER every 3 days.
[01:31] <Keybuk> ah, the openssl school of ABI stability
[01:37] <\sh> i followed the debian package :) which is replacing libquicktime1
[01:38] <\sh> and only a couple of packages ... easy one
[01:42] <hunger> Is it normal for udev to not log anything?
[01:43] <Keybuk> udevd?
[01:43] <Keybuk> it doesn't log much, no
[01:44] <hunger> Keybuk: I did grep -r udev /var/log/*. Nothing shows up at all.
[01:44] <Keybuk> what were you expecting?
[01:44] <Keybuk> it only logs errors by default
[01:44] <Keybuk> otherwise it'd quite capably fill your hard drive
[01:45] <hunger> Keybuk: Dunno.
[01:45] <hunger> Keybuk: Some indication that it is doing something?
[01:45] <Keybuk> udevmonitor
[01:45] <Keybuk> (if you have the package on my disk :p)
[01:46] <infinity> HELPFUL.
[01:46] <hunger> Keybuk: Maybe I am expecting too much.
[01:48] <Keybuk> you could recompile it with DEBUG enabled ... then change the logging priority to debug
[01:49] <Keybuk> if you change it to info, you at least get an "message received" log message and logs of the children forked
[01:49] <Keybuk> (change the udev_log= line in /etc/udev/udev.conf, then restart udevd)
[01:49] <hunger> Keybuk: Thanks.
[01:51] <hunger> Keybuk: "Normal" daemons report when they were started/stopped and I was expecting something like that.
[01:51] <Keybuk> udevd tends to be started wayyyyy before syslogd
[01:51] <Keybuk> and never stops
[01:51] <hunger> Keybuk: Good point:-)
[01:52] <Keybuk> in fact, syslogd is started so late in the boot process, it's funny
[01:52] <hunger> Keybuk: It is not "normal" at all, so I shouldn't expect normal behaviour (for the varying values of "normal" you get in a unix env).
[01:52] <hunger> Keybuk: Yeap... startup sequence is a science of its own...
[01:54] <Mithrandir> Keybuk: any idea how I can get bootchart to survive init re-execing itself?  It appears that it just dies, for some reason.
[01:54] <Keybuk> Mithrandir: no idea why it would die
[01:54] <Keybuk> unless init deliberately kills it
[01:55] <Mithrandir> it might be sending a signal to the process group, but shouldn't a double & fix that?
[01:55] <Keybuk> if something's sending TERM to all processes, that would explain it going away :)
[01:55] <Keybuk> sadly not, it starts *before* init, remember
[01:55] <Mithrandir> I'm ignoring TERM, iirc.
[01:55] <Keybuk> and gets reparented and stuff once init starts
[01:55] <Mithrandir> not in the d-i setup
[01:56] <hunger> Hmmm... udev reports "error: unknwon bus" (if run with priority "info", does not show up with loglevel "error":-(
[01:56] <Mithrandir> there it's started by rootskel
[01:56] <Keybuk> oh, no idea about the d-i setup
[01:59] <tepsipakki> anyone have an idea why portmap is started on every runlevel but stopped only on runlevel one?
[02:00] <Keybuk> tepsipakki: bet it's also stopped in rc0 and rc6
[02:01] <tepsipakki> nope
[02:01] <Keybuk> cute
[02:01] <tepsipakki>  update-rc.d portmap start 43 S . start 32 0 6 . start 18 2 3 4 5 . stop
[02:01] <tepsipakki> 81 1 .
[02:01] <tepsipakki> but there are bugs open..
[02:01] <tepsipakki> so nevermind
[02:02] <tepsipakki> I've just shoved networking out of rcS and it seems to work fine
[02:02] <tepsipakki> there's way too much done in rcS...
[02:02] <Keybuk> uh, don't do that
[02:03] <Keybuk> you'll break dhcp
[02:03] <Keybuk> S:S40networking brings up the loopback interface, without that dhclient goes nuts
[02:03] <tepsipakki> where is dhcp started?
[02:03] <Keybuk> via hotplug
[02:03] <tepsipakki> hotplug-net?
[02:04] <Keybuk> kinda
[02:04] <Keybuk> dhcp is started from the hotplug net.agent when the card is plugged in
[02:04] <Keybuk> on boot, the S:S40hotplug script (via net.rc) fakes events for cards already plugged in at boot
[02:04] <tepsipakki> I moved ifupdown, networking, hotplug-net, portmap, mountnfs.sh, ntpdate
[02:04] <Keybuk> those are held until S:S41hotplug-net because of dhclient needing S:S40networking to run first
[02:05] <Keybuk> you forgot S:S40pcmcia ;)
[02:05] <Keybuk> and buggered anyone with /usr or /home on an NFS partition
[02:05] <infinity> Keybuk : Oh, pshaw, no one does THAT.
[02:06] <tepsipakki> keybuk: well.. are those needed for su-mode?-)
[02:06] <Keybuk> "su-mode" ?
[02:06] <tepsipakki> init 1
[02:06] <Keybuk> init 1 is kinda pointless on Ubuntu
[02:07] <tepsipakki> not for servers
[02:07] <Keybuk> true
[02:07] <Keybuk> but then you probably want networking and nfs on a server
[02:07] <Lathiat> its usefull if you forget yoru password?
[02:07] <Lathiat> or X went spang?
[02:07] <Lathiat> doesnt work so well with initrds ?
[02:08] <Keybuk> Lathiat: happens after the initrd/initramfs has happened
[02:08] <Keybuk> you can use "break" to be dumped inside the initramfs, that's quite fun
[02:08] <Lathiat> passing init-/bin/sh executes after the initramfs now?
[02:08] <Keybuk> always has
[02:08] <Lathiat> was it always like that?
[02:08] <Lathiat> im sure i used to end up in initrds 
[02:08] <Lathiat> with a shell
[02:09] <Keybuk> not sure what it did with initrd
[02:09] <Keybuk> I think it still left you in the real filesystem though
[02:10] <sivang> Keybuk: if you have the package you wanted me to test (yeasterday's convo) I can try it now (re: udevplugd)
[02:10] <Keybuk> sivang: can you just do as root ... for uevent in $(find /sys -name uevent); do echo "add" > $uevent; done
[02:13] <sivang> Keybuk: sec
[02:13] <sivang> Keybuk: I did that, what should I be seeing?
[02:14] <sivang> Keybuk: hmm, find /sys -name uevent gives nothing
[02:14] <Keybuk> you running 2.6.15?
[02:15] <sivang> Keybuk: err, no. my dapper dist-upgrades didn't do that. Or is it a private kernel package from your personal page?
[02:15] <Nafallo> will 2.6.15 be default soon? :-)
[02:15] <Kamion> it's not the default kernel in dapper yet, and won't be until all of this is sorted out
[02:15] <Nafallo> ah
[02:16] <Kamion> sivang: linux-image-2.6.15-<whatever> in dapper
[02:16] <Kamion> Nafallo: if you want to make it happen faster, you could try Keybuk's test
[02:16] <Keybuk> sivang: is in the archive, but not default
[02:16] <sivang> Keybuk: ok, I will fetch it and retry
[02:16] <Nafallo> sure, where is it? :-)
[02:16] <Keybuk> Nafallo: that line of shell ^^
[02:17] <sivang> Keybuk: hmm, no -smp anymore to the package name?
[02:18] <Kamion> sivang: no, see changelog
[02:18] <Kamion>      YES! I know there are no i386 SMP packages. The 686 and k7 packages are
[02:18] <Kamion>      SMP enabled. In fact, they are also UP enabled "WHAT!?!? HUH!?!". Yes, SMP
[02:18] <Kamion>      on i386 now has the added benefit of killing lock ops dynamically on boot.
[02:18] <Kamion>      This is still in testing phase. I also need to fix a few things to get it
[02:18] <Nafallo> Keybuk: that for-thingie? :-)
[02:18] <Kamion>      working for modules.
[02:19] <Keybuk> Nafallo: yes
[02:20] <Keybuk> Kamion: oh, and other random 15-ish bug heads-up ... there's some new bug in the order that psmouse and usbmouse get loaded, where you can end up with non-functioning touchpads, etc.
[02:21] <Nafallo> Keybuk: and then? :-)
[02:22] <Nafallo> reboot and watch the chassi turn red? :-)
[02:22] <Keybuk> Nafallo: you're still here ... that's all I needed to know
[02:22] <Nafallo> lol
[02:23] <Keybuk> ... I still can't get used to the new bling-tastic aptitude
[02:25] <sivang> Keybuk: bling-tastic?
[02:25] <Treenaks> bling-a-rific!
[02:25] <Mithrandir> sivang: blingbling
[02:26] <sivang> Mithrandir: ?
[02:26] <sivang> Kamion: what's the UP enabled thing?
[02:26] <Kinnison> sivang: fancy, over-the-top, pretty
[02:26] <sivang> Kinnison: thanks god for you :)
[02:27] <Kinnison> sivang: "bling" often referrs to over-the-top jewelery
[02:27] <Keybuk> sivang: it keeps making "suggestions" about how to resolve problems
[02:27] <Keybuk> it's a bit mad
[02:28] <sivang> Keybuk: heh
[02:29] <sivang> Keybuk: anyway, I'm installed with 2.6.15 , now let's reboot
[02:29] <Kamion> sivang: UP == uniprocessor
[02:29] <Kamion> don't worry about it
[02:30] <sivang> Kamion: k, thx
[02:44] <sivang> Keybuk: hrm, that kernel doesn't boot for me 
[02:45] <sivang> "BUG: Soft lockup detected on CUP#0" . uspalsh is corrupted completely, and even rescue mode doesn't boot (overlooked the error)
[02:46] <ogra> sivang, usplash is known
[02:46] <infinity> sivang : Complain to BenC on #ubuntu-kernel, please.
[02:46] <infinity> ogra : usplash corrupting doesn't account for the kernel BUG.
[02:46] <ogra> yup
[02:47] <sivang> infinity: will do
[03:02] <siretart> debian seems to be able to request 'binNMU's when only a rebuild is needed. In the past, we reuploaded with 'buildX' suffix in the version string. To we keep that or can we request rebuilds from the buildd admins?
[03:04] <segfault> is there any "ubuntu samba team"?
[03:04] <Kamion> siretart: we're probably going to keep our current scheme at least for now.
[03:05] <Kamion> binNMUs have existed on a haphazard basis in Debian for a long time, but it's only very recent that the release team's been able to request systematic rebuilds on all architectures; and I don't think that mechanism will exist in Launchpad at least to begin with, so there's not much point rushing to switch over to it now
[03:05] <siretart> segfault: I don't think so
[03:05] <Kinnison> binNMUs are ugly as sin too
[03:06] <Kamion> in any case the buildX scheme distributes effort among more people and is thus better; the only reason Debian doesn't do that is because the maintainer lock is more prevalent there
[03:06] <siretart> Kinnison: whats wrong with them? I try to understand whats bad about the idea
[03:06] <siretart> Kamion: I see. thanks for explanation
[03:06] <Kamion> siretart: it's technically unpleasant to associate binNMUs with the correct source package version at the moment, especially retroactively
[03:07] <Kamion> (you need to do that for GPL compliance etc.)
[03:07] <siretart> aaah, I see
[03:10] <Kamion> siretart: seriously though, do you really want to go from being able to just upload something to having to request that a buildd admin do something for you? :)
[03:11] <mjg59> ogra: Yeah. It looks rather less featureful...
[03:13] <magnon> Simira: ping
[03:13] <Simira> magnon : you live
[03:13] <magnon> Simira: laptop broken down :(
[03:13] <magnon> been handicapped for a week
[03:14] <Simira> magnon : hurrah. Don't you have other computers?
[03:14] <magnon> actually no
[03:14] <siretart> Kamion: I was rather thinking about the DD's, who might be confused about 'patches' from ubuntu, which only have a different debian/changelog
[03:14] <magnon> I'm living in my bag these days
[03:15] <Kamion> siretart: any Debian developer worth the name should not be overly confused by a changelog that says "Rebuild for <such-and-such a reason>. No source changes."
[03:15] <Simira> magnon : ah, right. so where are you now?
[03:15] <magnon> Simira: was just going to mention though, that I'm writing an open letter to the department of education and a debate to aftenposten, so someone in the department will read "edubuntu" in a few days
[03:16] <magnon> I'm dropping by my mother's, and I'll be heading down to venstres hus in a minute
[03:18] <Keybuk> pitti: fix dhcp, kthxbyue
[03:20] <Keybuk> you derooted it a little too hard, it can't write to its own state files
[03:26] <pitti> Keybuk: oh, it worked fine after I derooted it some months ago
[03:26] <pitti> must have broken it during merge...
[03:28] <Keybuk> /var/run/dhclient*leases are owned by root
[03:28] <Keybuk> dhclient runs as dhcp, so cannae write to 'em
[03:33] <pitti> Keybuk: erm, why /var/run? Last time the leases were stored in /var/lib/dhcp3
[03:33] <pitti> $ l /var/lib/dhcp3/
[03:33] <pitti> insgesamt 12
[03:33] <pitti> -rw-r--r-- 1 root  root  439 2005-11-17 07:24 dhclient.leases
[03:33] <pitti> -rw-r--r-- 1 dhcpd dhcpd 640 2005-11-23 15:16 dhcpd.leases
[03:36] <Keybuk> hmm
[03:36] <Keybuk> Nov 23 14:32:47 localhost dhclient: can't create /var/run/dhclient.ath0.leases: Permission denied
[03:37] <Lathiat> shouldnt that be dhclient/ath0.leases, or soemthing?
[03:38] <Keybuk> (it's also worth pointing out that your own dhclient.leases there is owned by root <g>)
[03:39] <Keybuk> *giggle*
[03:40] <Keybuk> pitti: it might help if ifup didn't override the default on the command-line, eh?
[03:40] <pitti> Keybuk: heh, yes
[03:41] <Keybuk> though why is your dhclient.leases owned by root?  smells fishy
[03:41] <Keybuk> mine is too (in /var/lib/dhcp3)
[03:41] <pitti> Keybuk: it's already several days old; might have been an intermediate product from the merging
[03:41] <pitti> -rw-r--r-- 1 root  root  438 2005-11-23 15:16 /var/run/dhclient.eth0.leases
[03:41] <pitti> that's the current one
[03:41] <pitti> hm, why is it owned by root?
[03:42] <Keybuk> I'll leave you to fix that, shall I? :p
[03:42] <pitti> yep
[03:43] <pitti> $ ps h -o euser,ruser,suser,egroup,rgroup,sgroup -C dhclient3
[03:43] <pitti> 101      101      101      101      101      101
[03:43] <pitti> *boggle*
[03:51] <pitti> Keybuk: *headdesk* IZ ifupdown bug
[03:51] <pitti> Keybuk: sudo dhclient eth0 writes into /var/lib/dhcp3; sudo ifup eth0 writes into /var/run/dhclient.eth0
[03:52] <pitti> Keybuk: and what happens with /var/run at boot?
[03:52] <pitti> right, it's cleared
[03:52] <Keybuk> isn't it nice when a bug comes into sharp focus
[03:54] <pitti> Keybuk: btw, root is ok, that was deliberate. I drop root privs only after opening the pid file, for exactly this reason
[03:54] <pitti> so, open file, drop root, do network stuff unprivileged
[03:55] <Keybuk> uh...
[03:55] <Keybuk> that means you've written the file as root
[03:55] <Keybuk> which means you can't rewrite the file later
[03:55] <Keybuk> it closes the file after writing it, it doesn't keep it open and trunc it
[03:55] <Keybuk> so the next time it tries to write the updated lease, it'll fail?
[03:56] <pitti> Keybuk: gotta check; at least the server keeps the file handle open for this reason
[03:56] <Keybuk> the client definitely doesn't
[03:56] <Keybuk> it's an fopen call that's failing ;)
[03:57] <pitti> ah, I see it
[03:57] <pitti> it closes and opens again, how stupid
[03:58] <pitti> k, great; however, that does not seem to explain #18148
[03:58] <pitti> that guy has the problem after a boot
[03:58] <Keybuk> right, 18148 is explained by /var/run getting wiped on boot :)
[03:58] <Keybuk> I found the permission denied error on my system, then remembered having read that bug -- rather than the other way around
[03:58] <Keybuk> so
[03:59] <Keybuk> 1) dhclient can't write over its own leases file
[03:59] <pitti> (at second and further attempts)
[03:59] <Keybuk> yeah, which was my bug
[03:59] <Keybuk> 2) ifup overrides dhcp's path and tells it to put the lease in /var/run, which is wiped on boot
[03:59] <Keybuk> so 18148 isn't #1, it's #2
[04:00] <pitti> right, I agree
[04:00] <pitti> 1) is easy to fix, will do that
[04:16] <\sh> woohoo...debhelper is installable again :) thx
[04:32] <siretart> \sh: which mirror do you use?
[04:35] <Mithrandir> Keybuk: can you tell me my why init thinks it's a bright idea to kill stuff when reading the inittab?
[04:37] <\sh> siretart: archive.ubuntu.com
[04:37] <Keybuk> Mithrandir: heh
[04:37] <Keybuk> because init is stupid
[04:38] <Mithrandir> Keybuk: well, apart from that?
[04:38] <Keybuk> I, for one, welcome our new init-ng overlords
[04:38] <\sh> elmo: please sync gcipher gconf geda-gschem gmetadom from unstable, overriding ubuntu changes ok
[04:38] <\sh> elmo: thx :)
[04:39] <sivang> siretart: even from here (.il) it seems that a.u.c has the best and fastest routing I found. Hence I'm also using it.
[04:39] <Keybuk> *blink*
[04:39] <Keybuk> From: 	Jiaqing Du <jqdu@163.com>
[04:39] <Keybuk> To: 	scott@netsplit.com
[04:39] <Keybuk> Subject: 	2004/planet--devel--0.2
[04:39] <Keybuk> ...
[04:39] <Keybuk> I think he meant "baz" not "mail"
[04:39] <infinity> Kamion : Ping.
[04:40] <Mithrandir> elmo: please sync libttf-alee, overriding ubuntu changes is ok.
[04:40] <siretart> sivang: seems to be a problem with my apt-proxy
[04:40] <siretart> strange.. still
[04:40] <elmo> sh: nothing to sync for gmetadom
[04:40] <elmo> sh: otherwise done
[04:41] <infinity> elmo : Is libgtk+2.0-directfb-dev asking to be promoted to main, perchance?  cdebconf appears to want it now.
[04:41] <elmo> Mithrandir: -EPKG
[04:41] <\sh> elmo: gmetadom hm?
[04:41] <Mithrandir> elmo: EICANTSPEL, ttf-alee is the package name.
[04:41] <elmo>   gmetadom |    0.2.3-4 |        dapper | source
[04:42] <pitti> infinity: hit the guy who is responsible for approving the reports *cough*
[04:42] <\sh> ah...
[04:42] <\sh> elmo: sorry...but it was on our list..
[04:42] <pitti> infinity: you hit me so hard, now I can't fix cupsys any more
[04:43] <elmo> Mithrandir: done
[04:44] <Mithrandir> elmo: thanks
[04:46] <infinity> pitti : Perhaps I'm doing you a favour, then.
[04:46] <pitti> infinity: indeed you do. this stupid thing drives me up the wall since yesterday
[04:50] <sivang> infinity: did you had anything to do with that page btw?
[04:51] <Keybuk> Chris Halls started that iirc
[04:51] <infinity> sivang : I don't think I've ever even read it. :)
[04:52] <infinity> But I'm apparently quoted.  Neat.
[04:52] <mvo> at the top even!
[04:52] <siretart> hi mvo 
[04:52] <Keybuk> funny, I hadn't pegged infinity as a top
[04:52] <sivang> very very good, especially for me => http://women.alioth.debian.org/wiki/index.php/English/BuildingWithoutHelper
[04:53] <sivang> Keybuk: hehe
[04:53] <sivang> Keybuk: btw, me and BenC still trying to make 2.6.15 boot on my hardware :-)
[04:54] <siretart> mvo: the changelog to apt tells something about a 'apt-secure' manpage, but I cannot find it in apt/dapper
[04:54] <siretart> mvo: I also think that 'apt-get source -t <dist>' still does not work. could there be some foo with the merge?
[04:55] <mvo> siretart: my dapper has a apt-secure(8) manpage
[04:56] <mvo> siretart: apt-get source -t $dist works only if you have both deb/deb-src for the dist in sources.list. it's a limitation of the current code
[04:56] <infinity> Oh, that reminds me...
[04:56] <siretart> mvo: aaah, this explains!
[04:56] <infinity> mvo : Can I get a -t-me-harder option that pins at > 1000, so I can do forced sidegrades without writing a preferences file? :)
[04:57] <siretart> YAY! first upload to main ACCEPTED! :)
[04:57] <pitti> siretart: congrats
[04:58] <siretart> :)
[04:58] <mvo> infinity: hm, interessting idea. you want it for "install", right? because it should work for "source" already, no?
[04:58] <infinity> mvo : More evil, I want it for dist-upgrade. :)
[04:58] <Keybuk> aptitude upgrade "~i~Dsnappy" or whatever the silly thing is
[04:59] <mvo> infinity: *garr*
[04:59] <infinity> mvo : "apt-get -t-me-harder breezy dist-upgrade" should do a forced sidegrade of, say, sarge to breezy. :)
[04:59] <Keybuk> siretart: you know that the ACCEPTED things are automatic right? :)  we don't have elves checking incoming packages ... well, just NEW ones, and Colin doesn't like it when you call him and elf
[04:59] <mvo> infinity: woah, what is the use-case for this?
[04:59] <infinity> mvo : I already do this by writing out a pin in apt_preferences, would just be nice to skip that step.
[04:59] <sivang> Keybuk: lol
[04:59] <infinity> mvo : The use case is my own insanity.  Not sure if there's a better one. :)
[05:00] <infinity> mvo : But for install targets too, it makes sense.  -t pins at 990, so can't force downgrades.
[05:01] <siretart> Keybuk: hehe, got it
[05:01] <mvo> infinity: let me have a look
[05:01] <infinity> mvo : I could make an argument for the 'install' use case being useful.  The dist-upgrade use case is just me being lazy.
[05:01] <elmo> infinity: it's a new source+binary promotion to main, even if it is just a library, I don't think it classifies as obvious - I'd also like Colin to sign off that we actually want  it
[05:02] <infinity> elmo : Yeah, if you noticed in scrollback, right after I pinged you, pitti noted that he had an inclusion report to sign off on. ;)
[05:02] <Keybuk> infinity: you know you can do (aptitude reinstall "~i~Asnappy") right ... which does exactly what you want?
[05:02] <infinity> Keybuk : If I used apititude, I might know that. :)
[05:03] <pitti> infinity, elmo: no, I haven't
[05:03] <infinity> (I figure if apt is trying to provide that basic functionality though, it may as well make it work)
[05:03] <Keybuk> aptitude is the home of all strange and crackful things you might want to do with your machine
[05:03] <infinity> pitti : Oh, there isn't a report for it
[05:03] <infinity> pitti : ?
[05:03] <Mithrandir> Keybuk: strange it doesn't include a MTA or an editor yet.
[05:03] <pitti> infinity: no, nobody wrote one
[05:03] <pitti> infinity: of course I can do it myself
[05:03] <infinity> pitti : Oh.  I'll harass Kamion, then.
[05:03] <pitti> depending on how urgent it si
[05:03] <pitti> s/si/is/
[05:04] <infinity> pitti : It's only urgent if we like the installer to be buildable.
[05:04] <Mithrandir> pitti: depends, I guess.  afaik, we need it for all the gui shiny bits.
[05:05] <mvo> infinity: -o APT::Force-Default-Release-At-All-Cost=breezy? (may "Force-Default-Release" is enough)
[05:06] <infinity> mvo : That's not much less typing that setting up a preferences file. :)
[05:06] <mvo> infinity: slacker ;) 
[05:06] <Mithrandir> infinity: it can be tab-completed.  That's a win.
[05:06] <infinity> mvo : Hey, it's a (very pointless) wishlist anyway, feel free to ignore me.
[05:06] <infinity> mvo : I have other wishlists that are higher on my list of things to bribe you for.
[05:07] <infinity> mvo : Like the "orig + diff from different mirrors" wishlist.
[05:07] <pitti> infinity: at least apt-get install package/distro works and downgrades
[05:07] <Keybuk> pitti: do you know off the top of your head when dhclient puts its pid file by default, or doesn't it?
[05:08] <pitti> Keybuk: well, right at the beginning, just before dropping privs
[05:08] <mvo> infinity: orig+diff from different mirrors?
[05:08] <infinity> pitti : Ahh, kay, then it's just the dist-upgrade thing I'm missing, which is a pretty insane wishlist.
[05:08] <doko> \sh: we finally can start the libstdc++ allocator change, I'm preparing a current list of packages.
[05:08] <Keybuk> pitti: so I've just worked out why ifupdown changes it
[05:08] <Keybuk> pitti: the dhclient defaults don't include the interface name
[05:09] <\sh> doko: rock :)
[05:09] <doko> mvo: ^^^ please could you take care of the apt dependent packages?
[05:09] <pitti> Keybuk: right
[05:09] <Keybuk> pitti: which buggers laptop users ;)
[05:09] <pitti> Keybuk: but why does it change the dir?
[05:09] <mvo> doko: yes
[05:09] <Keybuk> it didn't change the dir ... you changed it, or upstream changed it
[05:09] <Keybuk> it must've been /var/run once
[05:09] <Keybuk> and ifupdown never got fixed
[05:09] <pitti> ah, I see
[05:10] <pitti> Keybuk: I think at the time when I derooted dhclient, I did not dare to change ifupdown (seemed too intrusive to me, too close to release)
[05:10] <pitti> Keybuk: so now would actually be a good time, to see what breaks with the new path
[05:10] <infinity> mvo : Debian #164399
[05:11] <Keybuk> pitti: yeah
[05:11] <Keybuk> what was the bug#?
[05:12] <pitti> Keybuk: mine? 18148?
[05:12] <Keybuk> ok
[05:12] <pitti> Keybuk: unless you see an apparent reason why some program might rely on /var/run, I can fix it now
[05:13] <Keybuk> I've already fixed it :)
[05:13] <pitti> ah, great
[05:13] <mvo> infinity: http://paste.ubuntulinux.nl/4917 (not sure this is in general interessting enough for all users though :)
[05:13] <infinity> mvo : For people with local mirrors that update once a day, but archive.ubuntu in sources.list (to stay fresh for development), the use case for that bug is obvious.
[05:14] <mvo> infinity: I haven't looked at the new one yet, my comment was about the --force-default-release
[05:14] <Keybuk> I hate nowebm
[05:14] <Mithrandir> infinity: it's only an issue for bandwidth-limited continents.  That is, .au and lamont. :-)
[05:14] <infinity> mvo : I see that.  You work fast. ;)  Patch looks sane.
[05:14] <infinity> Mithrandir : Thpt.
[05:15] <siretart> hm. iptables still not building all modules, appearently..
[05:15] <pitti> Mithrandir: or, rather, all European states but Norway :)
[05:15] <mvo> infinity: that was any easy one, I wish I could come up with something as easy for your build-dep on .dsc files request :/
[05:15] <Mithrandir> pitti: I haven't heard fabbione complain much about lack of bandwidth. :-)
[05:15] <mvo> infinity: or making the acquire interface (a bit) more sane 
[05:16] <infinity> mvo : build-dep on .dsc would be sweet, but I'm pretty resigned to the reality that it will never happen.
[05:16] <infinity> mvo : I may just tear the coce out of sbuild that does build-dep handling, and use it as a wrapper for apt-get.
[05:16] <infinity> s/coce/code/
[05:16] <mvo> infinity: would you accept a external helper for it? my gdebi code is more or less cabable of doing it 
[05:17] <infinity> I think we spell that "G-I-V-I-N-G U-P"
[05:17] <siretart> E: Wow, you exceeded the number of versions this APT is capable of.
[05:17] <siretart> :/
[05:17] <Kamion> infinity: pong
[05:17] <Mithrandir> infinity: /usr/lib/pbuilder/pbuilder-satisfydepends might be able to?
[05:18] <siretart> mvo: how many number of versions does apt currently handle?
[05:18] <infinity> Kamion : Yo, do you intend to do an inclusion report for libgtk+2.0-directfb-dev (on stealthily promote it when noone's looking)?
[05:18] <Kamion> infinity: well. If we want to support the graphical installer at all, then we need it. OTOH I doubt that will be ready for dapper, so maybe I should just compile it out for now ...
[05:18] <infinity> Mithrandir : I'd prefer to reimplement sbuild's handling, actually, since it's considered the "correct" way (given that it will give the same results as the buildds)
[05:19] <infinity> Mithrandir : pbuilder and sbuild have been known to disagree.
[05:19] <infinity> mvo : I have enough ways to work around the .dsc build-dep thing for now that it's not a big deal.  We can revisit it some other time.
[05:20] <infinity> mvo : Like when I have to rewrite sbuild in python for launchpad.
[05:20] <Keybuk> didn't someone already do that?
[05:20] <infinity> mvo : The Debian bug I aimed you at would earn you a case of beer, though.  And I also suspect it's the worst of all my wishlist bugs.
[05:21] <mvo> siretart: hm, how many versions did you managed to produce? how many dists in your sources.list?
[05:21] <infinity> Keybuk : LP is using the Perl sbuild for now.  A proper reimplementation still needs love and testing.
[05:21] <mvo> infinity: ok
[05:21] <infinity> Keybuk : And it's hardly critical, since sbuild "just works", and we can tear it out and replace it at any point without disrupting the rest of the world.
[05:22] <Keybuk> fsvo works, but yes
[05:22] <Keybuk> I'd always assumed that the vaunted "launchpad buildds" were not just running sbuild
[05:22] <siretart> mvo: I reduced it now, I had 6, which is obviously too much
[05:22] <Keybuk> it makes it seem so trivial for so much work that went into them
[05:23] <siretart> (and no, this is NOT my /etc/apt/sources.list)
[05:23] <mvo> siretart: interessting that you didn't got a mmap-out-of-range error first
[05:23] <siretart> mvo: I did, but that I could increase
[05:23] <infinity> Keybuk : Well, there's no buildd to be found, just sbuild for the actual package builds.
[05:23] <mvo> siretart: heh :)
[05:23] <infinity> Keybuk : Everything above that layer is a complete rewrite and re-think.
[05:26] <siretart> GRRR. I hate marillat :/
[05:29] <mako-pumpkin> mjg59: you around?
[05:29] <Simira> mako-pumpkin : halloween is over, you know
[05:31] <Mithrandir> who turned mako into a pumpkin?
[05:31] <mjg59> mako-pumpkin: Hi
[05:32] <Simira> cinderella's fairie god-mother?
[05:32] <HiddenWolf> mjg59, you just said Hi to a pumpkin...
[05:32] <mjg59> HiddenWolf: Yes...
[05:32] <HiddenWolf> Talking to pumpkins is generally not a good sign, you know.
[05:32] <mjg59> It talked to me first, and I've seen mako do stranger things
[05:33] <Mithrandir> mjg59: please don't elaborate on that.
[05:33] <dilinger> bippity boppity boo
[05:33] <mako-pumpkin> mjg59: nicholas negroponte is going to get a new laptop.. and it's going to run ubuntu
[05:33] <mako-pumpkin> mjg59: what will it be
[05:33] <Mithrandir> mako-pumpkin: Thinkpad.
[05:33] <mjg59> mako-pumpkin: What does he want?
[05:33] <mjg59> Thinkpads are a good bet
[05:33] <Simira> zepto
[05:33] <Simira> :D
[05:34] <Mithrandir> Simira: they're problematic wrt linux.
[05:34] <dilinger> mako-pumpkin: between sunday and today, i haven't changed my mind about my x40 :)
[05:34] <mako-pumpkin> something that works out of the box under ubuntu.. probably x40 size i guess
[05:34] <mako-pumpkin> i like the x40
[05:34] <mako-pumpkin> but wondered if there was anything else
[05:34] <mjg59> mako-pumpkin: The X40 works out of the box and is X40 size
[05:34] <Mithrandir> mako-pumpkin: x40 is love.
[05:34] <Simira> Mithrandir : a guy claimed to Rupert in Nordialog that they were ok with most linux.
[05:34] <Mithrandir> Simira: I can claim that pigs can fly.
[05:34] <mako-pumpkin> well, that's what i had said already :)
[05:34] <Simira> Mithrandir : they do
[05:35] <Simira> well, one of them, anyway
[05:35] <sivang> mjg59: you'd be happy to know that T43/p's work out of the box , including sleep
[05:35] <mjg59> sivang: I'd bloody well hope so, we tested the damn things
[05:35] <mjg59> :)
[05:35] <sivang> mjg59: it's so SWEET , believe me
[05:36] <sivang> mjg59: very nice beast machines in a very neat 14" pack
[05:36] <sivang> anyway,  I have to reboot to try get 2.6.15 working in this place
[05:37] <slomo> sivang: good luck :)
[05:37] <mako-pumpkin> mjg59: a favorite from the hp machines?
[05:38] <ogra> mako-pumpkin, !
[05:38] <mjg59> mako-pumpkin: I've only had stuff from the 14" range
[05:38] <mjg59> The 6220 is nice, but worse than a T43 in pretty much every respect
[05:38] <ogra> mako-pumpkin, you didnt give a vote for mhz yesterday ...
[05:38] <mjg59> Same for the 4200
[05:40] <infinity> Thinkpads are a hard religion to leave.
[05:40] <pitti> aaaaarrgh /me beats cupsys to death
[05:41] <Keybuk> if there was a 12" Thinkpad with a touchpad, I'd've gone for it
[05:41] <siretart> touchpads are overrated
[05:41] <Keybuk> I don't rate them at all ... I depend on them
[05:41] <Keybuk> either that, or a real mouse
[05:41] <infinity> The touchpad on my T43 is disabled, that's how much I love them.
[05:49] <mjg59> Oops.
[05:49] <mjg59> I just ctrl+alt+backspaced the machine I ran X from, rather than the machine X was running on
[05:49] <dilinger> i always liked touchpads, but the clit on my x40 is much nicer than the ones i've previously used.  i don't really miss the touchpad at all.
[05:56] <\sh> cu later ...going home
[06:32] <doko> dholbach: you could start change the library names of the *mm* packages
[06:32] <dholbach> doko: will do that
[06:55] <lamont-away> Kamion: ping
[07:32] <\sh> infinity / lamont-away: all packages which were synced/uploaded, since the broken debhelper/gcj, will be given-back automatically to the buildds?
[07:32] <lamont-away> \sh: they'll all be given back, yes
[07:32] <lamont-away> or have been
[07:33] <\sh> lamont-away: thx for the info :)
[07:37] <infinity> Will be, not have been.
[07:38] <infinity> There will be a mass give-back either in an hour or so (if I manage to stay awake), or when I wake up tomorrow.
[07:38] <infinity> (Not quite ready yet)
[07:41] <seb128> Diziet: "gtk2xtbin.h:44:27: error: X11/Intrinsic.h: No such file or directory"  (current firefox build)
[07:42] <infinity> seb128 : Again?
[07:42] <Diziet> Hello.
[07:42] <seb128> infinity: seems so
[07:42] <seb128> Diziet: hi :)
[07:42] <Diziet> Um.
[07:42] <Diziet> It built for me.
[07:43] <infinity> Diziet : You dropped my changes from 1.4.99+1.5rc2.dfsg-1ubuntu2
[07:43] <infinity> Diziet : (added build-dep on libxt-dev)
[07:43] <Diziet> Oh.  That wasn't there when I started :-/.
[07:43] <Diziet> Sorry about that.
[07:43] <infinity> No big deal.
[07:43] <Diziet> Do you want to put it back or shall I ?
[07:43] <infinity> Go nuts.  I'm babysitting half the world elsewhere, and it's approaching am, so I'm cranky. :)
[07:44] <Diziet> OK.
[07:44] <infinity> s/am/6am/
[07:47] <\sh> guys...what is the general policy regarding libcurl-ssl-dev? should it be replaced with libcurl3-gnutls-dev or libcurl3-ssl-dev?
[07:49] <infinity> \sh : If in doubt, use -ssl-dev (since that's the equivalent of the old -dev package), if the license reuiqres it, use gnutls-dev.
[07:50] <infinity> Some day, the gnutls variant will be feature-complete, and we may switch wholesale (and drop the libssl variant), but that's not quite feasible right now)
[07:51] <elmo> cron.daily's disabled for a while; shout if that's massively inconvenient
[07:51] <\sh> infinity: when I'm reading now the buildlogs and the debian packages are build-depending on libcurl-ssl-dev, the buildd is taking the first options which is gnutls-dev
[07:52] <\sh> infinity: and i'm unsure if or if not I should change those build-deps
[07:54] <infinity> elmo : Rather, actually.
[07:55] <infinity> elmo : How long is a while?
[07:55] <infinity> elmo : (I was just about to abuse wanna-build by forgetting half the world and wating for .daily to put everything back in needs-build)
[07:55] <infinity> waiting, too.
[07:56] <infinity> \sh : If the packages appear to work linked with the gnutls variant, no reason to change it, I guess.
[07:57] <\sh> k
[07:57] <elmo> infinity: till I fix the archive inconsistencies that are hindering LP
[07:57] <infinity> elmo : Ahh, fair enough.
[07:58] <elmo> anyway, I've run into a big problem with denise, so I'll re-enable it in a bit with the first round of changes
[07:58] <infinity> elmo : I'll just perpetrate my evil and go to sleep, assuming all will be well later on.
[07:58] <Diziet> infinity: Should be fixed now.  Sorry about that.
[08:09] <\sh> doko: are u fixing boost? :)
[08:10] <\sh> oh it's a give back
[08:10] <jdong_> siretart requested backport of apt... looks good to me; any reason not to do it?
[08:10] <\sh> jdong_: lets wait for the new version from mvo
[08:12] <siretart> \sh: which new version? whats bad with the current one?
[08:12] <siretart> \sh: I backported the current version of apt myself, and it worked for me [tm] 
[08:12] <\sh> siretart: apt needs a rebuild anyways...and mvo is bringing in more hot stuff
[08:13] <siretart> bah, I don't need more hot stuff. I need the new features of apt which are in dapper now :)
[08:13] <jdong_> \sh: ETA on "new version"?
[08:13] <\sh> siretart: then go :) I just annotated :)
[08:14] <jdong_> oh yes, FF 1.5, how will it be handled?
[08:14] <siretart> jdong_: nobody but mvo has
[08:14] <jdong_> is security team gonna do backports of 1.5?
[08:14] <\sh> jdong_: dunno..but I think soon
[08:14] <jdong_> since it seems like Mozilla.org is discontinuing security support for the 1.0 series
[08:14] <doko> \sh: should be already synced
[08:14] <siretart> jdong_: well, for FF1.5, there is a first version in dapper, which broke everything depending on firefox
[08:14] <jdong_> siretart: yeah, I know that
[08:15] <jdong_> I've already backported 1.5rc3 from dapper for myself :)
[08:15] <jdong_> yes, epiphany and friends need recompiles
[08:15] <\sh> doko: yeah...but it's not installable..so a give back..forget my question :)
[08:15] <siretart> jdong_: I'd suggest waiting a bit to see how quickly ff stabilizes. If we have a somewhat usable FF, we can think about backporting
[08:16] <jdong_> siretart: ok, but what does security team (pitti) have planned?
[08:16] <jdong_> I don't want to trample over them again
[08:16] <jdong_> need I bring back memories of mirrormax firefox? ;)
[08:16] <infinity> We'll have to evaluate the situation if and when upstream actually kills 1.0 security aupport.
[08:16] <infinity> support, too.
[08:17] <\sh> doko: btw...can I bother u again? regarding packages which we renamed in breezy (cxx trans) but debian didn't 
[08:17] <ajmitch> \sh: I say we change to match debian's naming, providing we have a clean upgrade from breezy
[08:17] <\sh> doko: how should we handle those merges? especially when debian upstreams orig package does work without any problems
[08:18] <siretart> jdong_: they would want to avoid backporting ff 1.5 to breezy, if possible. 
[08:18] <jdong_> see ya guys
[08:18] <siretart> cu jdong_ 
[08:18] <jdong_> siretart: I'd expect so
[08:18] <jdong_> bye :)
[08:18] <\sh> ajmitch: I will have a look on the rdepends
[08:19] <ajmitch> \sh: I don't think there's too many that have different naming
[08:19] <elmo> Kamion: around?
[08:19] <\sh> ajmitch: i just catched cal3d :)
[08:20] <\sh> where this situation appeared
[08:20] <ajmitch> \sh: most of it is due to poor communication with debian :)
[08:20] <ajmitch> we caught a few during breezy
[08:20] <\sh> ajmitch: most of it is because of the different naming convention
[08:21] <ajmitch> which we could have caught by talking with the debian maintainers :)
[08:21] <ajmitch> but it's a big job
[08:21] <siretart> doko: so the c2a transition has started? you wanted to write an email to ubuntu-devel iirc
[08:22] <ajmitch> siretart: there is one, isn't there?
[08:22] <\sh> ajmitch: yes..but there should be a "go go go" mail :)
[08:23] <ajmitch> send out the attack orders :)
[08:23] <siretart> ajmitch: I remember just an anoncement with a list of packages, and instructions to wait until further instructions
[08:23] <ajmitch> siretart: ah right
[08:23] <\sh> ajmitch: i'll started already :)
[08:23] <ajmitch> well I've really got to get ready for work ;)
[08:23] <infowolfe> does this channel handle kde stuff also?
[08:24] <infinity> Diziet : I hope your changelog was a typo..
[08:24] <\sh> infowolfe: kubuntu-devel
[08:24] <infowolfe> \sh, already there
[08:24] <infinity> Diziet : (it mentions "xft", but the missing build-dep was "libxt-dev")
[08:27] <mvo> jdong: eta for new apt: soon :)
[08:28] <siretart> mvo: too late ;)
[08:28] <mvo> siretart: already?
[08:28] <Kamion> elmo: not really ... is it something quick?
[08:28] <Kamion> lamont-away: pong, similarly
[08:28] <siretart> 20:18:16 -!- jdong_ [n=jdong@24.192.2.190]  has quit ["Lost terminal"] 
[08:29] <elmo> Kamion: no, never mind
[08:29] <lamont-away> Kamion: does 04:15 DCT sound like an OK time to do all 4 livecd builds (consecutively)
[08:29] <lamont-away> ?
[08:29] <siretart> DCT like in 'data center time'? ;)
[08:35] <Kamion> lamont-away: can I talk to you about that later? the answer is "maybe, will have to check the cdimage crontab closely", and I have to go to training
[08:35] <lamont-away> sure np
[08:38] <ajmitch> elmo: requesting sync of gnue-common, dropping ubuntu changes
[08:41] <jdong> mvo, others: can someone e-mail/IRC me when this new apt gets in?
[08:45] <siretart> I need to do stuff with gnupg in python, and I am looking for an gnupg module for python. As pyme does not seem to be maintained any longer, are there alternatives?
[08:45] <\sh> jdong: u can read it on dapper-changes :)
[08:45] <Keybuk> \o/
[08:45] <Keybuk> ./udevplug  0.07s user 0.02s system 76% cpu 0.124 total
[08:45] <\sh> python-gnugpinterface 
[08:45] <Keybuk> ... I think that should be fast enough
[08:46] <\sh> siretart: u got it?
[08:49] <sivang> Keybuk: what are you trying to make faster? hardware detection?
[08:49] <Keybuk> sivang: no, the kernel deals with hardware detection
[08:51] <Keybuk> I'm making hardware activation faster :)
[08:51] <Keybuk> ie. replacing udevstart and hotplug
[08:51] <jdong> Keybuk: doesn't that largely depend on how long the modprobe command takes?
[08:52] <siretart> \sh: /me checks
[08:52] <Keybuk> jdong: no
[08:52] <\sh> siretart: it's a gnupg to python pipe lib :) and installed by default...I will use it for glpbugs
[08:53] <siretart> \sh: mmmh. I need to get the fingerprint of a clearsign'ed file (in this case: a signed changes file)
[08:53] <siretart> \sh: is this possible with gpg at the cli?
[08:54] <Keybuk> we can't speed that up, but we can reduce the overhead of doing them
[08:54] <\sh> siretart: uh...dunno
[08:54] <doko> siretart: stuck for moderation
[08:54] <siretart> doko: ah. I see
[08:56] <doko> \sh: we should get the same package names as debian has. there were a few cases, where Debian didn't rename (and we did it wrongly). which packages are you thinking of?
[08:56] <\sh> cal3d 
[08:56] <\sh> the merge dropped our changes (renaming) and now it's also on the new allocator list :)
[08:57] <\sh> so if there is no "c2" suffix and it's new upstream, i don't rename the package...
[09:01] <siretart> \sh: FYI, it is possible (--with-fingerprint and --verify)
[09:01] <\sh> siretart: cool
[09:02] <sivang> bah
[09:02] <sivang> cp: writing `/boot/grub/menu.lst~': No space left on device
[09:05] <doko> \sh: just renamed it again
[09:06] <\sh> doko: well..debians package doesn't have a c2 ..and there are only 2 or 3 rdepends 
[09:07] <doko> \sh: the number of rdepends doesn't matter at all
[09:07] <\sh> doko: ok..I'll rename it then back to our naming convention and include the renaming of the new trans
[09:09] <tseng> jdong: up for some backport requests?
[09:09] <jdong> tseng: e-mail em to the list... I'm running out for an appointment
[09:10] <jdong> snowy today, gotta leave a bit earlier
[09:13] <doko> \sh: why rename it back???
[09:14] <siretart> \sh: please, lets stick to the package names the DD's chooses if in doubt
[09:34] <\sh> doko_: thx for the mail
[10:07] <seb128> mdz: could you give editbugs to eric@extremeboredom.net ?
[10:07] <mdz> seb128: have they read HelpingWithBugs?
[10:11] <seb128> mdz: dunno if he read the pages, but he points bugs to change/close on #ubuntu-bugs for like 1 hour and he does that right
[10:11] <mdz> seb128: ok, I will mail him
[10:11] <seb128> thanks
[10:15] <\sh> doko: u listed as well the packages which has to be rebuild, right? I can't see anjuta as a lib :)
[10:19] <doko> \sh: because it does have a shlibs file
[10:21] <\sh> autogenerated then
[10:41] <\sh> elmo: please sync c2hs from unstable, dropping ubuntu changes ok, thx
[10:54] <LaschW> jbailey: I'm getting a segmentation fault message booting linux-image-2.6.12-10-k7
[10:55] <LaschW> jbailey: 'sudo dpkg-reconfigure linux-image-2.6.12-10-k7' don't fix the problem.
[10:55] <jbailey> LaschW: Is that dapper?
[10:55] <LaschW> jbailey: Yepp
[10:55] <jbailey> When did it last work?
[10:56] <LaschW> I just installed the 2.6.12-10-k7 image. 2.6.12-9-k7 works fine
[10:57] <jbailey> Does it give you any output before the first segfault?
[10:57] <jbailey> And is it any different if you remove 'quiet' from the grub command line?
[10:57] <LaschW> jbailey: fabbione told me that his kernel image is ok and told me to ask you
[10:58] <LaschW> jbailey: Last message is uncompressing Linux... OK booting ....
[10:58] <LaschW> jbailey: After that there are 10 lines with segmentation fault
[10:58] <LaschW> jbailey: and then line 11 says: '0'
[10:58] <jbailey> Try adding the word panic to the kernel command line.
[10:59] <jbailey> See if that gives you a shell.
[11:00] <LaschW> jbailey: The system is booting after this messages. Thats what is a bit irritating to me
[11:00] <jbailey> Ah, that's annoying.
[11:01] <jbailey> Ummm, try panic, and see if you get any segfault messages before that.
[11:01] <jbailey> Then failing that, do so again with break
[11:01] <jbailey> We need to isolate down to a piece to start looking in to see what's segfaulting.
[11:01] <LaschW> jbailey: OK, give me a 1/4 hour...
[11:02] <jbailey> LaschW: Also, FWIW, I'm no longer the right person to ask.  Keybuk and Infinity are generally better choices.
[11:11] <mdke> BenC, around?
[11:13] <\sh> cleaning up my open bugzilla bugs..grmpf
[11:23] <LaschW> jbailey: Back again.. Have a look at: http://paste.ubuntulinux.nl/4935
[11:24] <jbailey> Right, so the first one is panic, the second one is with break?
[11:25] <LaschW> jbailey: the first one is with 'panic' option, the second one is my regular boot
[11:25] <jbailey> 'kay, so nothing earlier than 'panic', which is good.
[11:29] <HiddenWolf> LaschW, how do you get that output?
[11:29] <HiddenWolf> thought boot wasn't logged...
[11:30] <Surak> Hello
[11:30] <LaschW> HiddenWolf: The old fashioned way: Pancil, paper and writing it by hand :-))
[11:30] <jbailey> HiddenWolf: Serial console?
[11:30] <jbailey> Ouch. =)
[11:30] <HiddenWolf> LaschW, how many reboots did that take?
[11:31] <LaschW> HiddenWolf: some of them, not more than 4 till now.
[11:31] <HiddenWolf> ouch. :)
[11:31] <Surak> seb128_ : there?
[11:31] <jbailey> I've been known to take pictures of my screen. =)
[11:31] <seb128_> Surak: sort of, about what?
[11:31] <HiddenWolf> I really need to figure out the serial console trick
[11:31] <Surak> gstreamer's debug packages :-)
[11:31] <LaschW> As we like to say in Germany 'Zu bld ist nur wer sich nicht zu helfen weiss'
[11:32] <Surak> Do you remember that they wasn't compiling? You told me you were going to create debug packages.
[11:32] <HiddenWolf> LaschW, My german is too rusty, what does it say?
[11:33] <Surak> seb128_: there was some assembler stuff going wrong.
[11:33] <LaschW> HiddenWolf: word by word translated 'too stupid is who don't know how to help himself'
[11:34] <HiddenWolf> LaschW, ah, I see it now. :)
[11:34] <\sh> phew
[11:34] <seb128_> Surak: hum, right
[11:34] <sivang> \sh: how do you send a msg to all channel you're in in irssi?
[11:34] <HiddenWolf> sivang, /notice ?
[11:35] <HiddenWolf>  /notice chan message
[11:35] <sivang> HiddenWolf: thanks
[11:35] <\sh> sivang: i don't do those things
[11:35] <LaschW> jbailey: Any hint how I may figure out which process segfaults?
[11:35] <Surak> seb128_ : in http://bugzilla.ubuntu.com/show_bug.cgi?id=19434 you asked me for a debug backtrace. But as you were the one which would supply me the debug packages, I'm asking again ;-)
[11:36] <sivang> Surak: https://wiki.ubuntu.com/DebuggingProgramCrash
[11:36] <sivang> Surak: there you can find out how to create a debug package for yourself :)
[11:36] <\sh> dideldi..xterm-207 will hit the archives soon
[11:36] <seb128_> Surak: yeah, sorry, I've like 900 bugs to my list, not easy to keep everything uptodate
[11:36] <Surak> sivang: gstreamer packages wasn't compiling cleanly. 
[11:37] <sivang> Surak: ah, look better next time before you jump up to people with suggestions :)
[11:37] <sivang> anyway, I'm out for some sleep. Good Night all
[11:37] <Surak> sivang: so I posted this here on this channel some days ago, and seb128 told me
[11:37] <Surak> that he would provide me the packages, because it was weird stuff.
[11:38] <sivang> Surak: ah  I see, well sorry again.
[11:38] <Surak> sivang: but thanks for the link (even if I already was there )
[11:39] <sivang> Surak: np...
[11:39] <Surak> night sivang
[11:41] <\sh> gnarf
[11:42] <\sh> daniels: ping...what is now the correct position of app-defaults? /etc/X11/app-defaults or /usr/lib/X11/app-defaults?
[11:42] <\sh> we have two actually
[11:47] <daniels> \sh: /etc/X11/app-defaults
[11:47] <daniels> \sh: long-term, however, upstream are moving to the defaults in /usr/share/X11/app-defaults and local customisations in /etc/X11/app-defaults
[11:47] <daniels> \sh: what do you mean 'we have two'?
[11:48] <\sh> daniels: we have as well a lot of stuff installed in /usr/lib/X11/app-defaults
[11:48] <jbailey> LaschW: You're on that system right now?
[11:48] <\sh> or it's not cleaned correctly by dist-upgrade from breezy to dapper
[11:48] <LaschW> jbailey: Yepp
[11:48] <daniels> \sh: there shouldn't be stuff in /usr/lib/X11/app-defaults
[11:49] <daniels> \sh: willing to bet it's the random apps that got packaged, though; can you give me a list?
[11:49] <\sh> Xman Xmessage XSm
[11:49] <\sh> sure
[11:49] <daniels> mmm
[11:49] <daniels> yep, that'd be it
[11:49] <\sh> no
[11:49] <\sh> it's more
[11:49] <daniels> can you please file a bz entry, assign it to me?
[11:49] <\sh> daniels: of course..my pleasure :)
[11:49] <daniels> yeah, I've got a pretty good idea of which packages are doing it now
[11:49] <jbailey> LaschW: As root:  mkdir /tmp/foo; cd /tmp/foo; zcat /boot/initrd.img | cpio -i; cd ..
[11:49] <jbailey> LaschW: chroot foo
[11:49] <\sh> daniels: against what package?
[11:50] <jbailey> LaschW: Then you'll be in the initramfs encironment.
[11:50] <\sh> xserver-xorg?
[11:51] <mdke> i am jinxed, neither of my laptops have working dri in dapper, even though they've had it since hoary
[11:52] <\sh> daniels: i filed it against xserver-org, just because it's the global name for me :)
[11:53] <\sh> sabdf1?
[11:53] <daniels> \sh: yeah, that'll do, as long as I can see it on my bug list
[11:54] <\sh> daniels: it's assigned to u :)
[11:55] <\sh> and xterm upstream changed a lot in the standalone package
[11:55] <\sh> wow
[11:55] <LaschW> jbailey: Ok, done. What's next?
[11:56] <Surak> seb128: so, what should I do? the compile line is: cc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../libavcodec -DHAVE_AV_CONFIG_H=1 -fomit-frame-pointer -Wall -Wno-switch -msse -g -Wall -O0 -MT libmmxsse_la-fdct_mmx.lo -MD -MP -MF .deps/libmmxsse_la-fdct_mmx.Tpo -c fdct_mmx.c  -fPIC -DPIC -o .libs/libmmxsse_la-fdct_mmx.o
[11:56] <Surak> seb128_ : and the error is /tmp/ccq5gJtf.s: Assembler messages:
[11:56] <Surak> seb128_ : /tmp/ccq5gJtf.s:1996: Error: suffix or operands invalid for `pshufw'
[11:56] <Surak> seb128_ : make[9] : ** [libmmxsse_la-fdct_mmx.lo]  Erro 1
[11:57] <seb128_> Surak: may try to build the current dapper package?
[11:57] <Surak> is dapper still using gstreamer0.8?
[11:58] <jbailey> LaschW: Try a couple of the commands in /bin
[11:58] <jbailey> See if they all are broken for you (in which case klibc is probably hosed)
[11:58] <Surak> seb128_ : I'll try to buld the dapper package in breezy. Thanks for your attention.
[12:01] <seb128_> np, thanks for trying to get a backtrace
[12:01] <seb128_> yeah, dapper is using the same gst packages, out of the ffmpeg one
[12:01] <LaschW> jbailey: uname, cat, ln, true, mkdir are working
[12:01] <HiddenWolf> seb128_, when will it transition to gstreamer0.9?